Posts by T. F. M. Reader 1196 publicly visible posts • joined 19 Dec 2012
I came to quote the same sentence and ask what prevents the CA legislators to actually mandate this for all, obviating any need for age verification.
Yes, yes, I know GOOG, FB, et al. are in CA, no need to shout... Mind you, they won't like age verification very much, either.
In any case - upvoted.
the SFTW weekend column has been axed
What? Is that just because I've missed the significance of El Reg's survey ("what do you value most" and stuff like that) and have not marked NSFWSFTW as the most valuable feature (at least after the parting of Verity, Lewis, and Tim) on a par with the BOFH??
Whoever makes decisions at El Reg, do restore Dabbsy to his usual glory and glow (oh, that's Mme D., but still...), or I'll grass you to Simon and Stephen... and it will be "just a matter of time, really."
set up the first app to trigger not just one but a whole host of app services in one go
systemd?
applications in the future would be a collection of plug-ins
microservices?
"Subscribe to us because we have huge gaping holes in our functionality which can only be filled by subscribing to other apps!"
Which will enrich AMZN, GOOG, and MSFT by sending lots of data between various clouds, regions, and zones, while collecting and storing all sorts or private data in various elastic buckets for later abuse and reselling.
Cynical? Moi?
One relevant question is whether 4096/8192/more bit RSA will still be safe or using radically different algorithms will be the only way forward.
But then, as far as I understand, much of the strength of the quantum-resilient algorithms lies in the size of their keys...
a test to prove you are competent enough to vote
There is a famous gedankenexperiment on the topic. Pick a contentious issue, formulate a yes/no question, run 2 referendums: one among the general population, another among a large number (1K?, 10K? whatever) of the most intelligent, the most competent, the best educated, the most reasonable people chosen by the most appropriate method (details don't really matter).
Possible outcomes:
1. Same answer in the 2 referendums -> intelligence, competency, education, and reason make no difference. One possible explanation: maybe the people are not so dumb after all? Another: maybe those most suitable ones are not so exceptional? There are other possibilities, too.
2. The answers are different -> the above qualities do matter, but get lost in the democratic process. Now, who do we trust to devise the selection rules to disenfranchise all but the most suitable? The same people who voted differently from the majority of the population on the divisive issue in the experiment?
Difficult. Should we run a referendum on which outcome is better?
OMG! A lisper and a schemer saying sensible things about programming languages here! And mentioning Erlang, too! And I thought I was the last one to remember Greenspun's tenth rule... I am not alone!
Wait... Am I supposed to read something into the fact that both hide behind AC masks??? What have I done?
You pay only for getting your data back, assuming you have no other recourse, e.g., restoring from backups. [Aside: getting you data back is an assumption. IIRC Colonial pipeline paid, but the decryptor didn't work very well.]
It seems to me that this is a small part of dealing with a ransomware attack. Surely cleaning up and identifying how the criminals got into your network and plugging all the holes and making sure no warez remain after cleanup, including in backups, etc., is bound to be a much higher cost to the organization compared to the ransom payment itself.
So is the "I'll pay again" statement completely reasonable in the context of "I'll have much bigger problems if I am hit again" or is it completely unreasonable in the context of "restoring from backups is bloody hard but still just a small part of what I'll have to go through if I am hit again"?
Can one get insurance to cover the ransom payment? If I can I won't care about paying the next time - I will have already paid my premium. That's until the premium grows enough after the Nth breach... Can I get insurance to compensate for the downtime while dealing with the breach, as above? The premium may be a lot higher than just for the ransom - maybe better invest in actual security: teams, tools, procedures, equipment, etc?
No, I have not made any actual calculations. Hence all the question marks.
My late mother used to have really long phone conversations with her lady friends who were often retelling stories or situations of interest. My mom's part of the conversation (the only one I overheard) typically consisted of "Yes... Yes... Yes... Yes..." ad infinitum. For hours. I assume it was symmetric when it was my mom's turn to regale her friends with interesting stuff.
I guess I shouldn't write up my childhood memories in Google Docs...
Excess mortality is, indeed, the only parameter that makes sense in the context. Anything else will be affected by differences in testing, counting, transparency, etc., including "of Covid" vs. "with Covid" and what not.
Having said that, the material in the link pulls my eyebrow up. Annual data from 2015-2019? No way you can get significance from a sample of 5. (Aside: their "P-score" is not the p-value, which is the probability that the observed effect could happen by chance. You cannot get low p-value, and thus a high confidence that there is an effect, from a sample of 5) Besides, why use death counts? Population changes, too - use death rates (per 100K, per million, whatever - I found this normalization important, see below).
Next comment: "worst mortality for 8 years" means either "worst mortality over a very short period of time" or "nothing very special" (if the time series is actually a lot longer).
I did a comparison of 2020 mortality for the country I live in (hit by Covid as everyone else, several lockdowns, 2020 saw no vaccinations anywhere in the world, etc.), based exclusively on published data from our equivalent of ONS. I checked normalized (by population) mortality for 2000-2019 against 2020 (had to wait till the final figures for the latter were published).
To my surprise, the 2020 mortality was low against the previous 20 years - lower than either the mean or the median of the control sample. I struggle to see how lockdowns could save many lives to offset presumed Covid-related excess deaths. E.g., annual deaths in traffic accidents are low here (maybe ~20 times lower than deaths officially attributed to Covid in 2020). 20 years seems to be a short enough period to discount advances in medical care as a significant confusion factor (the country has advanced medical care in general). That was, in fact, a factor in my decision to limit the time series.
2020 mortality turned out relatively high against the previous 10 years, but not the highest. I have no idea what caused, e.g., 2012 to be so bad here. Even if 2020 mortality is high compared to the previous 5 years (I have not checked) the result can't be significant
The exercise was done purely for curiousity's sake. My conclusion: mortality in 2020 was nothing particularly special in this particular neighbourhood in this millennium.
I assume anyone can repeat the exercise for, say, UK after trawling ONS for some more data.
Disclaimer: the above should not be construed to imply that Covid is not a very serious disease. It obviously is, for some groups of people at least. I offer no firm explanation for the (surprising, for me) result. One may hypothesize that the fact that the life expectancy of the patients most likely to die of Covid (old and vulnerable) was something like 3 months might have something to do with it, but that's no proof. I didn't do any age-adjusted analysis.
Is it though? The article hints of at least correlation, if not causal relationship, of the (alleged... khm...) fraud scheme with sackings. Could it be that older staffers not allowed to apply their skills to CAMSS (sounds really IBM-ish) but still bringing home bread and butter from sales of stuff IBM actually did rather well in the past, such as mainframes, were gotten rid of to keep the lid on?
I didn't know the Communist Manifesto dated back to the ancient Rome... Your assumption that the Manifesto devised original terms is disturbing...
Pulling tongue out of left cheek... However, even the parallel between the Roman census term (basically, proletarii were citizens without significant property to their name) and the working class under 19th century capitalism predates Marx by some margin.
Are you sure A41 has a removable battery?
Apart from that I think the A52s I had to get last fall also ticks all the boxes (I think the price was about $500 back then). Dual SIM, even, but no removable battery. Sigh.
So, has GitHub (that's MSFT, right?) applied any sanctions to the guy? In my opinion, it's beyond unprofessional, regardless of whether one agrees with his views or not. I'd expect him to be kicked out and get a lifetime ban or something, if it is technically possible.
As for all the developers, presumably some (many?) of them commercial companies, that pulled this stuff without stopping twice that's essentially a supply chain attack. A trivial one, and, one assumes, trivially preventable. IANAL, and I may be hopelessly naive, but I think some level of "duty of care" is expected to be exercised while producing stuff others use, even if it is "just software".
And as a related point: I am not a JS expert by any stretch, but I thought that the browser was supposed to cuddle any JS in some kind of sandbox. How come a random piece can do IPC or delete arbitrary files or drop files on the Desktop? Can anyone enlighten me?
And as an afterthought: does it even work on a Mac?
I spent a part of my career doing massively parallel supercomputers, though it was many years ago (I started a bit after ASCI White, actually). I will be grateful to the commentariat if anyone can point to M1 Ultra's LINPACK results which is how supers, including ASCI White, are measured (and what they are designed for - not for LINPACK per se but for solving partial differential equations and such).
ASCI White actually never achieved 12 TFLOPS in practice, IIRC. Nvidia GPUs did a few years ago, but again on different benchmarks - again, IIRC.
All I saw in my admittedly perfunctory search was that M1 Ultra beats Intel and AMD. It is not even clear to me how far behind (in TFLOPS) a many-thousand-core GPU augmented with several hundred tensor processors it is. From what I saw M1 Ultra has 20 CPU cores, 64 GPU corres, and a 32-core "neural engine", whatever that is.
The current fastest super is ~45,000 faster that ASCI White (single CPUs evolved a lot faster indeed), has 7,630,848 cores, more than 5PB of memory, and is, by my estimate, about 40% as efficient (in MFLOPS/W) as M1 Ultra, for its LINPACK MFLOPS. That's not bad for a system that is ~26860 faster. I am much more excited by this than by the new M1.
All of this is not to diss M1 at all. I play around with both (older) Intel and (newer) M1 Macs at work every day and M1s are noticeably cooler and faster than Intels when running long and heavy parallelized compilations. All I am saying is that if you want to compare to a massively parallel super you must count oranges, namely LINPACK, for fairness. The supers probably will not even be able to run Netflix... On the other hand, it is not clear if an M1 Ultra can run this in 12 hours.
Maybe M1 Ultra will be great at LINPACK, too. I have not yet seen the numbers though.
@ThatOne: "It's the mediocre ones who are dangerous"
I realize the context is professional mediocrity, but in this day and age I can't help recalling the words of a prominent Russia human rights persona on one V. Putin right after the latter took over from Yeltsin: "Putin was a Lt. Colonel in the KGB. You know what a Lt. Colonel in the KGB is? Absolutely nothing."
If there is an extended Godwin's law on these pages I'll get my coat... --->
I don't think (serious) Russian cyber-attacks ever originate from Russian networks. Judging from the words in the article the thinking may be to prevent Russian-based criminals and spies from accessing non-Russian botnets.
Not entirely sure this particular step can achieve the goal. And it does seem aligned with Mr. Putin's aspirations of disconnecting Russia from all things Western. But in the spirit of "send them back to the stone age"...
I reserve judgement.
RT News and Sputnik News are no longer available for download from the App Store outside Russia.
What's the rationale for still allowing those inside Russia? Shouldn't Putin's internal propaganda machine be disrupted as much as possible? Or are those installed by default on Russkies' iThings so that there is no point? Genuine question.
And we have disabled both traffic and live incidents in Apple Maps in Ukraine as a safety and precautionary measure for Ukrainian citizens.
Given the reputation of Apple Maps I suppose it'll really help Ukrainians to avoid head-on collisions with 40-mile-long columns of Russian vehicles. Wait, do Russian troops, at least senior officers, carry iThings? Maybe leave it on and as default for them? No, not really a genuine question, just looking for small ways to help ...
One gotcha that still catches people out of course is opening an attachment
Another gotcha that catches me out every time is printing from something like Office.
Open a file, maybe look at the contents for a bit, hit print, try to close the file, get a "Save changes?" prompt, panic: what the HELL have I changed and should I save the modified version or keep the old one?!?!?
Why printing is considered logically equivalent to editing is beyond me.
Would kicking Russian businesses (and government, of course) off AWS/Azure/GCP/etc. hurt where it counts? It is not exactly shutting down Internet in Russia - personal stuff will still be routed and news will still reach the population (until Mr. Putin shuts it down and maybe Mr. Biden increases funding for Radio Free Europe or something), I agree this is an important factor. Its the business services that will be affected and it just might hurt. And that infrastructure can't be replicated easily, if they can't buy servers it'll be more difficult still.
No, I don't know how much they are dependent on "the cloud", but the thought is entertaining.
Reading the article I get the impression that the questions were framed as "Would you mind sharing some personal info to get better health care?"
It is absolutely not clear to me if questions like "What would you prefer: sharing every minute detail about yourself or paying a few pennies per month to get the same level of user experience?" were included.
I've never - literally never - bought a computer with preinstalled Windows. For many years this meant that the computers came with FreeDOS. I admit I never played with it - just installed Linux over it. Still appreciated the fact that at least it was a solution for the vendor who could list a product "without OS".
It's only relatively recently that major vendors started providing machines with Linux preinstalled. I have a Dell desktop that came with Ubuntu. It was still listed as "without OS" just to make me chuckle. Ubuntu was therefore unexpected, and I treated it as I would treat FreeDOS - wiped it and installed another brand of the same OS, purely as a matter of personal preference and per original plan.
The ThinkPad I am typing this on came with FreeDOS though.
<Tips Hat>
As a physicist, I see enormous possibilities for making rubber-hose cryptanalysis totally useless: the more you know of the quantum key the less certain you will be of the decrypted content.
As a computer scientist, I realize that the same uncertainty principle will of course apply to Byzantine generals. But that's just another kind of uncertain failure, innit?
Coat, please - I am off to prepare my next grant application! ---->
Magic numbers are not reliable either.
Just the other week I found out how unreliable: it turned out that Apple's fat binaries and java class files use the same magic number with a hex character sequence that more than one person apparently found cute.
It looks like Apple discovered the cuteness earlier, but by the time Java people got into the game Apple didn't interest anyone so no one checked, I guess.
Details can be easily found on the interwebs or in the magic file (on Linux, but not on macOS, for some reason).
@MrBanana: even more staggering workflow effects are common on, say, macOS. With Mission Control (AppleSpeak for virtual desktops) especially.
Meeting comes up in the calendar or someone just calls you. You are, say, on a virtual desktop with the Teams window. You click on the link in the calendar invitation or on the Join button, as appropriate.
The display switches to a different virtual desktop, something flashes there, too fast for human eye. The display switches back to the original virtual desktop before you realise that something happened.
You need to figure out that the OS for some reason thought that a URL (or the button) should be handled by your browser. It is the browser that realises that the URL should be handled by Teams. If you know that then you know enough to switch to the virtual desktop where your browser is open, and you'll find another Teams window with the call (not joined yet!) there.
As for "something went wrong", "we have encountered a problem", and "reconnecting" - those are there all the time: leave teams alone for longer than 15 minutes, connect to or disconnect from VPN, just about anything causes inability to connect to calls, send/receive messages, messages arriving out of order, etc., etc.
Clippy: I see you are expecting that it should "just work", eh?
@Gene Cash
True, but for me there is also another reason: if something happens to my second factor (e.g., phone) and I need to reach my stuff I understand how to restore access, say, at work, where the sysadmin knows me personally. I don't understand how I can convince Google that I am me if they lock me out.
The article does not, IMHO, provide all the context for some quotes. I encourage the commentariat to click on the Project Zero link.
My first reaction: So Apple and Google deal with problems in 70 and 72 days on average, respectively. But for WerbKit the lag is 73 days. I call it "about average", not "an outlier".
After looking at the Project Zero page: 70 vs. 72 is for mobile phones only. For browsers, Apple's 73 "average days to repair" is a clear outlier (from a sample of 3 vendors). For the rest, the tables are quite interesting, and Apple don't look all that well (not the worst - there is always Microsoft and, well, Oracle...).
The sampling methodology of Project Zero is not that clear to me, but at least it helped bring my originally raised eyebrow back in its place.
May I suggest to MSFT's boffins to just release a reset-to-defaults Tay into the world again and see if it becomes nicer or ruder? Maybe tweak the algos a bit, giving more weight to more recent stuff.
That could be a better, or at least an alternative, indicator of what goes on, compared to surveys.
Back to the spring of 2020. Technology becomes a substitute for life, essentially. FB, ZM, TWTR, PINS et al. become the main/only way to communicate. NFLX and AMZN (Prime) stand for entertainment. The latter also becomes the main shopping/delivery option. Tech stocks, including all of those mentioned above, heat up.
2 years later it looks like some of us may be on the verge of getting out of house arrest. Pronto, FB, NFLX, AMZN, TWTR, PINS, ZM are back to the price levels of spring or early summer of 2020 (just checked the charts). Maybe some people actually prefer it on the outside rather commenting on El Reg wasting time in front of a screen, if allowed?
It will probably disappoint some, but maybe - just maybe? - it's a correction rather than collapse?
Trickle down economics doesn't work.
The article you link to says something very specific. It looks at trickle-down economics from the angle of cutting taxes to "the rich" (that includes companies, etc.). It says that trickle-down economics assumes that the savings will be invested in the economy one way or another and, well, trickle-down will occur. It then says that oftentimes the savings are not put back into economy, so the aforementioned assumption of trickle-down economics is, in general, unfounded.
All that is fine. In this case though, Mr. Bezos whom many love to hate has actually spent almost half a billion bucks. Some of it will, in fact, be paid as taxes, some will get transferred to various vendors up the shipbuilders' supply chain, salaries and bonuses will be paid, employees (and owners, and their families) will spend their income, shops and restaurants and pubs and service providers and employees thereof will get their share, etc.
So in this specific case the basic assumption of trickle-down economics is factually correct and Steve Button's comment was exactly on point. IMHO.
Jeopardy gives you an answer and you have to respond with the correct question.
I am dying of curiosity: has 42 ever been dealt on Jeopardy and what was the question? I tried to google, but the only vaguely relevant thing on the 1st page of results was about someone named Amy Schneider who apparently was a Jeopardy winner at some point and, possibly independently, was 42 years of age at some point. Doesn't quite cut it, IMHO...
half the DoD needs a new box
To be fair, I wouldn't be surprised if half the DoD only needed to reboot their boxes.
<rant>
Almost missed a customer session yesterday, was reminded by WhatsApp. The invitation was not in my calendar. An hour later got another, unrelated, invitation by mail, accept, check - not in the Calendar, either. Pattern recognition engaged and locked on target. Check OWA - not there, either. Open a ticket with IT... Think a bit... Reboot the computer for the first time in about 10 days on a creative whim - both invitations magically appear in the local Calendar and OWA. Cancel the IT ticket...
That's a rather new (<15 months old) MacBook Pro with the latest macOS - shouldn't need to be replaced even for USAF... Still no clue why rebooting should affect the server that OWA accesses... The case doesn't look all that different from needing an hour to send an email or to open Excel, but can't be fixed by a HW upgrade.
In my defence, I am not used to rebooting computers myself. Typing this on an 8 year old personal Linux laptop with uptime of more than 86 days... Got a new battery for it a couple of months ago, see no need for any other upgrade at this point.
</rant>
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
